Skip to content

ComputeSage StackBench MCP server

com.computesage/stackbenchcomputesage.comstreamable-httpv0.2.1Scanned

GPU and LLM inference benchmarks, hardware evidence, deployment recommendations, and launch configs.

5 tools: 4 change things and 1 only read.

Tools: 0 delete, 0 irreversible, 4 write, 1 read.
  • Reads private data · 0
  • Reads text others write · 0
  • Can send data out · 0
  • 1 finding

Tools listed without sign-in. Remote posture

With Anlyon

4 tools here change things. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the ComputeSage StackBench MCP server can do

The ComputeSage StackBench MCP server is published by computesage.com, a domain the MCP Registry verified. It has 5 tools. It can create or change data (4 tools: check_deployment_fit, predict_performance, recommend_deployment and generate_launch_config), but cannot delete data. 1 tool only reads.

None of its tools reads private data, brings in text other people write, or sends data out.

The fixed checks matched 1 finding (annotation-conflict). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 12, 2026, version 0.2.1.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

5 tools

Changes things 4Claude Code: ask
  • check_deployment_fit$0.01 USDC on Base mainnet: check model/configuration fit against bounded StackBench capacity evidence.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    $0.01 USDC on Base mainnet: check model/configuration fit against bounded StackBench capacity evidence.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • generate_launch_config$0.50 USDC: generate deterministic vLLM or llama.cpp launch templates from an exact StackBench recommendation without claiming a launch occurred.1 finding

    Why it is in this group

    • The name "generate_launch_config" says "generate": a word for a change.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    Findings

    • Warn

      The server marks "generate_launch_config" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "generate_launch_config" says "generate": a word for a change.

      What this check looks for
    Description, as the server wrote it
    $0.50 USDC: generate deterministic vLLM or llama.cpp launch templates from an exact StackBench recommendation without claiming a launch occurred.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • predict_performance$0.05 USDC: predict configuration throughput, memory, and power from bounded StackBench evidence with confidence, missing metrics, and caveats.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    $0.05 USDC: predict configuration throughput, memory, and power from bounded StackBench evidence with confidence, missing metrics, and caveats.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • recommend_deployment$0.25 USDC: recommend what to buy or run from bounded StackBench fit, performance, confidence, evidence, caveats, and economics only when supported.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    $0.25 USDC: recommend what to buy or run from bounded StackBench fit, performance, confidence, evidence, caveats, and economics only when supported.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
Only reads 1Show 1 read-only tool
  • search_evidence$0.00 free search of bounded public StackBench benchmark and failure evidence; use this for evidence-only requests.No findings

    Why it is in this group

    • The name "search_evidence" says "search": a word for reading.
    Description, as the server wrote it
    $0.00 free search of bounded public StackBench benchmark and failure evidence; use this for evidence-only requests.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or untrusted-input or outbound.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write0
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out0
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:36 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
4
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 4 tools that change things (4 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://mcp.computesage.com/mcp: tools/list returned 5 tools to an anonymous client, including check_deployment_fit (write), predict_performance (write), recommend_deployment (write), generate_launch_config (write)

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

1 warn
  • The server marks "generate_launch_config" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "generate_launch_config" says "generate": a word for a change.

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 0.2.1. Values in angle brackets are yours to fill in.

claude mcp add --transport http stackbench https://mcp.computesage.com/mcp
claude mcp add --transport http stackbench https://mcp.computesage.com/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__stackbench__search_evidence"
    ],
    "ask": [
      "mcp__stackbench__check_deployment_fit",
      "mcp__stackbench__predict_performance",
      "mcp__stackbench__recommend_deployment",
      "mcp__stackbench__generate_launch_config"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as stackbench. Review it before you use it.

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/com.computesage/stackbench/verify?hash=8432926403e3c335fa8c3c721a311c1eefe915829df7f7f498a2c325b136ab30" | jq -e '.changed == false'

Pinned hash 8432926403e3, from Nov 12, 2026, 22:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/com.computesage/stackbench/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for com.computesage/stackbench
[![Anlyon Radar](https://radar.anlyon.com/badge/com.computesage/stackbench.svg)](https://radar.anlyon.com/servers/com.computesage/stackbench)

Source: MCP Registry. Synced Oct 21, 2026, 17:53 UTC.

ComputeSage StackBench MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the ComputeSage StackBench MCP server safe?

No scan can promise that. What its definitions show: 5 tools, 4 of which can change, send or delete something, and 1 finding from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the ComputeSage StackBench MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the ComputeSage StackBench MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the ComputeSage StackBench MCP server have?

5 tools: 4 write, 1 read. check_deployment_fit, predict_performance, recommend_deployment, generate_launch_config and search_evidence.

Does the ComputeSage StackBench MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the ComputeSage StackBench MCP server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the ComputeSage StackBench MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 12, 2026.